Data-driven approaches to materials exploration and discovery are building momentum due to emerging advances in machine learning. However, parsimonious representations of crystals for navigating the vast materials search space remain limited. To address this limitation, we introduce a materials discovery framework that utilizes natural language embeddings from language models as representations of compositional and structural features. The contextual knowledge encoded in these language representations conveys information about material properties and structures, enabling both similarity analysis to recall relevant candidates based on a query material and multi-task learning to share information across related properties. Applying this framework to thermoelectrics, we demonstrate diversified recommendations of prototype crystal structures and identify under-studied material spaces. Validation through first-principles calculations and experiments confirms the potential of the recommended materials as high-performance thermoelectrics. Language-based frameworks offer versatile and adaptable embedding structures for effective materials exploration and discovery, applicable across diverse material systems.
23 August 2021 in Journal of Materials Chemistry A
We explore n-type doping of BiCuSeO, an otherwise well-known p-type thermoelectric material, and find halogens to be the most effective n-type dopants.
Doping remains a bottleneck in discovering novel functional materials for applications such as thermoelectrics (TE) and photovoltaics. The current computational approach to materials discovery is to identify candidates by predicting the functional properties of a pool of known materials, and hope that the candidates can be appropriately doped. What if we could “design” new materials that have the desired functionalities and doping properties? In this work, we use an approach, wherein we perform chemical replacements in a prototype structure, to realize doping by design. We hypothesize that the doping characteristics and functional performance of the prototype structure are translated to the new compounds created by chemical replacements. Discovery of new n-type Zintl phases is desirable for TE; however, n-type Zintl phases are a rarity. We demonstrate our doping design strategy by discovering 7 new, previously-unreported ABX4 Zintl phases that adopt the prototypical KGaSb4 structure. Among the new phases, we computationally confirm that NaAlSb4, NaGaSb4 and CsInSb4 are n-type dopable and potentially exhibit high n-type TE performance, even exceeding that of KGaSb4. Our structure prototyping approach offers a promising route to discovering new materials with designed doping and functional properties.
Alloying is a common technique to optimize the functional properties of materials for thermoelectrics, photovoltaics, energy storage etc. Designing thermoelectric (TE) alloys is especially challenging because it is a multi-property optimization problem, where the properties that contribute to high TE performance are interdependent. In this work, we develop a computational framework that combines first-principles calculations with alloy and point defect modeling to identify alloy compositions that optimize the electronic, thermal, and defect properties. We apply this framework to design n-type Ba2(1−x)Sr2xCdP2 Zintl thermoelectric alloys. Our predictions of the crystallographic properties such as lattice parameters and site disorder are validated with experiments. To optimize the conduction band electronic structure, we perform band unfolding to sketch the effective band structures of alloys and find a range of compositions that facilitate band convergence and minimize alloy scattering of electrons. We assess the n-type dopability of the alloys by extending the standard approach for computing point defect energetics in ordered structures. Through the application of this framework, we identify an optimal alloy composition range with the desired electronic and thermal transport properties, and n-type dopability. Such a computational framework can also be used to design alloys for other functional applications beyond TE.